In the dappled sunlit waters of Caribbean mangrove forests, tiny box jellyfish bob in and out of the shade. Box jellies are distinguished from true jellyfish in part by their complex visual system — the grape-size predators have 24 eyes. But like other jellyfish, they are brainless, controlling their cube-shaped bodies with a distributed network of neurons.
That network, it turns out, is more sophisticated than you might assume. On Friday, researchers published a report in the journal Current Biology indicating that the box jellyfish species Tripedalia cystophora have the ability to learn. Because box jellyfish diverged from our part of the animal kingdom long ago, understanding their cognitive abilities could help scientists trace the evolution of learning.
The tricky part about studying learning in box jellies was finding an everyday behavior that scientists could train the creatures to perform in the lab.
Anders Garm, a biologist at the University of Copenhagen and an author of the new paper, said his team decided to focus on a swift about-face that box jellies execute when they are about to hit a mangrove root. These roots rise through the water like black towers, while the water around them appears pale by comparison. But the contrast between the two can change from day to day, as silt clouds the water and makes it more difficult to tell how far away a root is. How do box jellies tell when they are getting too close?
“The hypothesis was, they need to learn this,” Dr. Garm said. “When they come back to these habitats, they have to learn, how is today’s water quality? How is the contrast changing today?”
In the lab, researchers produced images of alternating dark and light stripes, representing the mangrove roots and water, and used them to line the insides of buckets about six inches wide. When the stripes were a stark black and white, representing optimum water clarity, box jellies never got close to the bucket walls. With less contrast between the stripes, however, box jellies immediately began to run into them. This was the scientists’ chance to see if they would learn.
After a handful of collisions, the box jellies changed their behavior. Less than eight minutes after arriving in the bucket, they were swimming 50 percent farther from the pattern on the walls, and they had nearly quadrupled the number of times they performed their about-face maneuver. They seemed to have made a connection between the stripes ahead of them and the sensation of collision.
Going further, the researchers removed visual neurons from the box jellyfish and studied them in a dish. The cells were shown striped images while receiving a small electrical pulse to represent collision. Within about five minutes, the cells started sending the signal that would cause a whole box jellyfish to turn around.
“It’s amazing to see how fast they learn,” said Jan Bielecki a postdoctoral researcher at the Institute of Physiology at Kiel University in Germany, also an author of the paper.
Researchers who were not involved in the study called the results a significant step forward in understanding the origins of learning. “This is only the third time that associative learning has been convincingly demonstrated in cnidarians,” a group that includes sea anemones, hydras and jellyfish, said Ken Cheng, a professor at Macquarie University in Sydney, Australia, who studies the animals. “And this is the coolest demonstration, replete with physiological data.”
The results also suggest that box jellyfish possess some level of short-term memory, because they can change their behavior based on past experience, said Michael Abrams, a postdoctoral researcher at the University of California, Berkeley, who studies the neuroscience of jellyfish sleep. He wonders how long the box jellies remember what they’ve learned. If they are taken out of the tank for an hour and then returned to it, do they have to learn what to do all over again?
In future work, the researchers hope to identify which specific cells control the box jellyfish’s ability to learn from experience. Dr. Garm and his colleagues are curious about the molecular changes that happen in these cells as the animals incorporate new information into their behavior.
They wonder, too, whether the capacity to learn is universal among nerve cells, regardless of whether they are part of a brain. It might explain their peculiar persistence in the tree of life.
“There are organ systems popping up and going away all the time,” Dr. Garm said. “But nervous systems — once they are there, they very rarely go away again.”
Perhaps the ability to learn is one reason they are still here.